{smcl}
{com}{sf}{ul off}{txt}{.-}
      name:  {res}<unnamed>
       {txt}log:  {res}\\file\UsersW$\wrr15\Home\My Documents\My Files\TAXES AND ECONOMIC GROWTH IN OECD COUNTRIES - A META-ANALYSIS\SUBMISSION TO PFR\DATA AND PROGRAMS\Part1 Results(20200102).smcl
  {txt}log type:  {res}smcl
 {txt}opened on:  {res} 3 Jan 2020, 07:50:04
{txt}
{com}. set more off
{txt}
{com}. set type double 
{txt}
{com}. graph drop _all
{txt}
{com}. import excel "TAX.xlsx", sheet("Stata") firstrow case (lower)
{res}{text}(140 vars, 713 obs)

{com}. gen endog = (tsls == 1 | gmm == 1)
{txt}
{com}. 
. *---------------------------------------------------------------------*
. *   TABLE 4: Distribution of Tax Effects by Fiscal Policy             *
. *---------------------------------------------------------------------*
. // FULL SAMPLE
. summ coefficient if predneg == 1, detail

                         {txt}Coefficient
{hline 61}
      Percentiles      Smallest
 1%    {res}    -.854         -.9987
{txt} 5%    {res}    -.522          -.854
{txt}10%    {res}    -.463         -.5853       {txt}Obs         {res}        156
{txt}25%    {res}    -.336          -.575       {txt}Sum of Wgt. {res}        156

{txt}50%    {res}   -.1235                      {txt}Mean          {res}-.1785359
                        {txt}Largest       Std. Dev.     {res} .2076465
{txt}75%    {res}    -.036          .1671
{txt}90%    {res}    .0489           .175       {txt}Variance      {res} .0431171
{txt}95%    {res}     .124           .183       {txt}Skewness      {res}-.7983064
{txt}99%    {res}     .183           .247       {txt}Kurtosis      {res} 3.820552
{txt}
{com}. summ coefficient if predother == 1, detail

                         {txt}Coefficient
{hline 61}
      Percentiles      Smallest
 1%    {res}    -1.91          -3.52
{txt} 5%    {res}     -.58         -3.001
{txt}10%    {res}   -.3754         -2.022       {txt}Obs         {res}        507
{txt}25%    {res}     -.19          -1.98       {txt}Sum of Wgt. {res}        507

{txt}50%    {res}    -.063                      {txt}Mean          {res}-.0845357
                        {txt}Largest       Std. Dev.     {res} .7541614
{txt}75%    {res}     .002            2.7
{txt}90%    {res}     .055           4.14       {txt}Variance      {res} .5687593
{txt}95%    {res}      .14           5.74       {txt}Skewness      {res} 10.61044
{txt}99%    {res}    .9106          12.72       {txt}Kurtosis      {res} 175.7032
{txt}
{com}. summ coefficient if predpos == 1, detail

                         {txt}Coefficient
{hline 61}
      Percentiles      Smallest
 1%    {res}   -.6753         -.6753
{txt} 5%    {res}   -.3613           -.37
{txt}10%    {res}  -.23355         -.3613       {txt}Obs         {res}         50
{txt}25%    {res}    -.118         -.3526       {txt}Sum of Wgt. {res}         50

{txt}50%    {res}    .0265                      {txt}Mean          {res}  .031996
                        {txt}Largest       Std. Dev.     {res} .2478908
{txt}75%    {res}    .1435           .371
{txt}90%    {res}     .362          .3787       {txt}Variance      {res} .0614498
{txt}95%    {res}    .3787           .524       {txt}Skewness      {res} .2404782
{txt}99%    {res}    .8204          .8204       {txt}Kurtosis      {res} 4.693386
{txt}
{com}. 
. quietly summ coefficient, detail
{txt}
{com}. scalar low = r(p5)
{txt}
{com}. scalar high = r(p95)
{txt}
{com}. keep if coefficient > low & coefficient < high
{txt}(72 observations deleted)

{com}. 
. // TRUNCATED SAMPLE
. summ coefficient if predneg == 1, detail

                         {txt}Coefficient
{hline 61}
      Percentiles      Smallest
 1%    {res}    -.522          -.524
{txt} 5%    {res}   -.4652          -.522
{txt}10%    {res}    -.449         -.5153       {txt}Obs         {res}        146
{txt}25%    {res}      -.3          -.488       {txt}Sum of Wgt. {res}        146

{txt}50%    {res}   -.1205                      {txt}Mean          {res}-.1680116
                        {txt}Largest       Std. Dev.     {res} .1741409
{txt}75%    {res}    -.037           .124
{txt}90%    {res}     .029          .1266       {txt}Variance      {res}  .030325
{txt}95%    {res}     .076          .1274       {txt}Skewness      {res}-.4831863
{txt}99%    {res}    .1274          .1489       {txt}Kurtosis      {res} 2.111504
{txt}
{com}. summ coefficient if predother == 1, detail

                         {txt}Coefficient
{hline 61}
      Percentiles      Smallest
 1%    {res}     -.44           -.52
{txt} 5%    {res}     -.37         -.4906
{txt}10%    {res}    -.288           -.45       {txt}Obs         {res}        457
{txt}25%    {res}    -.174           -.44       {txt}Sum of Wgt. {res}        457

{txt}50%    {res}     -.06                      {txt}Mean          {res}-.0958776
                        {txt}Largest       Std. Dev.     {res} .1313941
{txt}75%    {res}  -.00313           .146
{txt}90%    {res}     .039           .146       {txt}Variance      {res} .0172644
{txt}95%    {res}     .076          .1623       {txt}Skewness      {res}-.8109335
{txt}99%    {res}      .14           .166       {txt}Kurtosis      {res} 3.211208
{txt}
{com}. summ coefficient if predpos == 1, detail

                         {txt}Coefficient
{hline 61}
      Percentiles      Smallest
 1%    {res}     -.37           -.37
{txt} 5%    {res}   -.3613         -.3613
{txt}10%    {res}   -.2474         -.3526       {txt}Obs         {res}         38
{txt}25%    {res}   -.1497         -.2474       {txt}Sum of Wgt. {res}         38

{txt}50%    {res}  -.01205                      {txt}Mean          {res}-.0439105
                        {txt}Largest       Std. Dev.     {res} .1449069
{txt}75%    {res}     .051          .1369
{txt}90%    {res}    .1369          .1428       {txt}Variance      {res}  .020998
{txt}95%    {res}    .1435          .1435       {txt}Skewness      {res}-.7170208
{txt}99%    {res}    .1452          .1452       {txt}Kurtosis      {res} 2.665649
{txt}
{com}. 
. *-------------------------------------*
. *   TABLE 6: Sample Means             *
. *-------------------------------------*
. summ ///
> g7 eu15 eumem oecd ///
> gdp pcgdp ///
> marginal differenced etr ///
> lrcase1 lrcase2 lrcase3 ///
> peerreviewed originalpubyear ///
> cs panel ///
> length originalmidyear ///
> ols gls endog ///
> seols sehet sehac ///
> income laggeddv countryfe investment tradeopenness human popgrowth ///
> employgrowth unemploymentrate inflation 

{txt}    Variable {c |}        Obs        Mean    Std. Dev.       Min        Max
{hline 13}{c +}{hline 57}
{space 10}g7 {c |}{res}        641    .1170047    .3216769          0          1
{txt}{space 8}eu15 {c |}{res}        641    .0639626    .2448773          0          1
{txt}{space 7}eumem {c |}{res}        641    .0312012     .173997          0          1
{txt}{space 8}oecd {c |}{res}        641    .7878315    .4091628          0          1
{txt}{space 9}gdp {c |}{res}        641    .2589704    .4384114          0          1
{txt}{hline 13}{c +}{hline 57}
{space 7}pcgdp {c |}{res}        641    .7410296    .4384114          0          1
{txt}{space 4}marginal {c |}{res}        641    .0904836    .2870974          0          1
{txt}{space 1}differenced {c |}{res}        641    .1716069    .3773328          0          1
{txt}{space 9}etr {c |}{res}        641    .9063963    .2915041          0          1
{txt}{space 5}lrcase1 {c |}{res}        641    .7020281    .4577243          0          1
{txt}{hline 13}{c +}{hline 57}
{space 5}lrcase2 {c |}{res}        641    .0530421    .2242925          0          1
{txt}{space 5}lrcase3 {c |}{res}        641    .2449298    .4303814          0          1
{txt}peerreviewed {c |}{res}        641    .6614665    .4735805          0          1
{txt}originalpu~r {c |}{res}        641    2006.952    5.374055       1993       2015
{txt}{space 10}cs {c |}{res}        641    .0093604    .0963704          0          1
{txt}{hline 13}{c +}{hline 57}
{space 7}panel {c |}{res}        641    .9906396    .0963704          0          1
{txt}{space 6}length {c |}{res}        641    31.37441    7.119092          5         40
{txt}originalmi~r {c |}{res}        641    1985.161     5.28115     1970.5     2004.5
{txt}{space 9}ols {c |}{res}        641    .6770671    .4679625          0          1
{txt}{space 9}gls {c |}{res}        641    .1544462    .3616581          0          1
{txt}{hline 13}{c +}{hline 57}
{space 7}endog {c |}{res}        641    .1684867    .3745903          0          1
{txt}{space 7}seols {c |}{res}        641    .5865835    .4928308          0          1
{txt}{space 7}sehet {c |}{res}        641    .2449298    .4303814          0          1
{txt}{space 7}sehac {c |}{res}        641    .1684867    .3745903          0          1
{txt}{space 6}income {c |}{res}        641    .5585023    .4969535          0          1
{txt}{hline 13}{c +}{hline 57}
{space 4}laggeddv {c |}{res}        641    .1669267    .3732016          0          1
{txt}{space 3}countryfe {c |}{res}        641    .8330733    .3732016          0          1
{txt}{space 2}investment {c |}{res}        641    .5850234    .4931028          0          1
{txt}tradeopenn~s {c |}{res}        641    .1700468    .3759673          0          1
{txt}{space 7}human {c |}{res}        641    .4399376     .496767          0          1
{txt}{hline 13}{c +}{hline 57}
{space 3}popgrowth {c |}{res}        641    .2433697    .4294515          0          1
{txt}employgrowth {c |}{res}        641    .3775351     .485149          0          1
{txt}unemployme~e {c |}{res}        641    .0904836    .2870974          0          1
{txt}{space 3}inflation {c |}{res}        641    .1310452    .3377134          0          1
{txt}
{com}. 
. *--------------------------------------------------------------------------------*
. *   TABLE 5: First Estimates of Tax Effects                                      *
. *--------------------------------------------------------------------------------*
. 
. // FIXED EFFECTS
. 
. // Generating transformed variables for FE
. gen feprecision = 1/se
{txt}
{com}. gen fetstat = coefficient/se
{txt}
{com}. gen prednegg = predneg/se
{txt}
{com}. gen predposs = predpos/se
{txt}
{com}. 
. // TABLE 5/PANEL A
. // NO CORRECTION FOR PUBLICATION BIAS
. //This regression gives equal weight to each estimate
. // TABLE 5 - Panel A/Column 1
. regress fetstat feprecision prednegg predposs, noc vce(cluster idstudy)

{txt}Linear regression                               Number of obs     = {res}       641
                                                {txt}F(3, 41)          =  {res}    11.64
                                                {txt}Prob > F          = {res}    0.0000
                                                {txt}R-squared         = {res}    0.1762
                                                {txt}Root MSE          =    {res} 2.6309

{txt}{ralign 78:(Std. Err. adjusted for {res:42} clusters in idstudy)}
{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26}    Robust
{col 1}     fetstat{col 14}{c |}      Coef.{col 26}   Std. Err.{col 38}      t{col 46}   P>|t|{col 54}     [95% Con{col 67}f. Interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 1}feprecision {c |}{col 14}{res}{space 2}-.0047615{col 26}{space 2} .0028139{col 37}{space 1}   -1.69{col 46}{space 3}0.098{col 54}{space 4}-.0104443{col 67}{space 3} .0009212
{txt}{space 4}prednegg {c |}{col 14}{res}{space 2}-.0377245{col 26}{space 2} .0088096{col 37}{space 1}   -4.28{col 46}{space 3}0.000{col 54}{space 4}-.0555158{col 67}{space 3}-.0199332
{txt}{space 4}predposs {c |}{col 14}{res}{space 2} .0280898{col 26}{space 2} .0089455{col 37}{space 1}    3.14{col 46}{space 3}0.003{col 54}{space 4} .0100239{col 67}{space 3} .0461556
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. lincom -prednegg + predposs

{p 0 7}{space 1}{text:( 1)}{space 1}{space 1}{res}- prednegg + predposs = 0{p_end}

{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 1}     fetstat{col 14}{c |}      Coef.{col 26}   Std. Err.{col 38}      t{col 46}   P>|t|{col 54}     [95% Con{col 67}f. Interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 9}(1) {c |}{col 14}{res}{space 2} .0658143{col 26}{space 2} .0119814{col 37}{space 1}    5.49{col 46}{space 3}0.000{col 54}{space 4} .0416174{col 67}{space 3} .0900111
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}

{com}. //This regression gives equal weight to each study
. // TABLE 5 - Panel A/Column 2
. regress fetstat feprecision prednegg predposs [pweight = weight], noc vce(cluster idstudy)
{txt}(sum of wgt is 38.11478993733051)

Linear regression                               Number of obs     = {res}       641
                                                {txt}F(3, 41)          =  {res}     9.48
                                                {txt}Prob > F          = {res}    0.0001
                                                {txt}R-squared         = {res}    0.1735
                                                {txt}Root MSE          =    {res} 2.4874

{txt}{ralign 78:(Std. Err. adjusted for {res:42} clusters in idstudy)}
{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26}    Robust
{col 1}     fetstat{col 14}{c |}      Coef.{col 26}   Std. Err.{col 38}      t{col 46}   P>|t|{col 54}     [95% Con{col 67}f. Interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 1}feprecision {c |}{col 14}{res}{space 2}-.0067066{col 26}{space 2} .0037241{col 37}{space 1}   -1.80{col 46}{space 3}0.079{col 54}{space 4}-.0142275{col 67}{space 3} .0008144
{txt}{space 4}prednegg {c |}{col 14}{res}{space 2} -.048044{col 26}{space 2} .0178482{col 37}{space 1}   -2.69{col 46}{space 3}0.010{col 54}{space 4}-.0840891{col 67}{space 3}-.0119988
{txt}{space 4}predposs {c |}{col 14}{res}{space 2} .0323796{col 26}{space 2} .0075044{col 37}{space 1}    4.31{col 46}{space 3}0.000{col 54}{space 4} .0172241{col 67}{space 3}  .047535
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. lincom -prednegg + predposs

{p 0 7}{space 1}{text:( 1)}{space 1}{space 1}{res}- prednegg + predposs = 0{p_end}

{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 1}     fetstat{col 14}{c |}      Coef.{col 26}   Std. Err.{col 38}      t{col 46}   P>|t|{col 54}     [95% Con{col 67}f. Interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 9}(1) {c |}{col 14}{res}{space 2} .0804235{col 26}{space 2}   .01867{col 37}{space 1}    4.31{col 46}{space 3}0.000{col 54}{space 4} .0427187{col 67}{space 3} .1181283
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}

{com}. 
. // TABLE 5/PANEL B
. // CORRECTION FOR PUBLICATION BIAS
. //This regression gives equal weight to each estimate
. // TABLE 5 - Panel B/Column 1
. regress fetstat feprecision prednegg predposs, vce(cluster idstudy)

{txt}Linear regression                               Number of obs     = {res}       641
                                                {txt}F(3, 41)          =  {res}    31.81
                                                {txt}Prob > F          = {res}    0.0000
                                                {txt}R-squared         = {res}    0.1101
                                                {txt}Root MSE          =    {res} 2.2216

{txt}{ralign 78:(Std. Err. adjusted for {res:42} clusters in idstudy)}
{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26}    Robust
{col 1}     fetstat{col 14}{c |}      Coef.{col 26}   Std. Err.{col 38}      t{col 46}   P>|t|{col 54}     [95% Con{col 67}f. Interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 1}feprecision {c |}{col 14}{res}{space 2}-.0004235{col 26}{space 2} .0005983{col 37}{space 1}   -0.71{col 46}{space 3}0.483{col 54}{space 4}-.0016318{col 67}{space 3} .0007848
{txt}{space 4}prednegg {c |}{col 14}{res}{space 2} -.022658{col 26}{space 2} .0049875{col 37}{space 1}   -4.54{col 46}{space 3}0.000{col 54}{space 4}-.0327305{col 67}{space 3}-.0125856
{txt}{space 4}predposs {c |}{col 14}{res}{space 2} .0378059{col 26}{space 2}  .005112{col 37}{space 1}    7.40{col 46}{space 3}0.000{col 54}{space 4} .0274821{col 67}{space 3} .0481297
{txt}{space 7}_cons {c |}{col 14}{res}{space 2}-1.555963{col 26}{space 2} .3114562{col 37}{space 1}   -5.00{col 46}{space 3}0.000{col 54}{space 4}-2.184962{col 67}{space 3}-.9269647
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. lincom -prednegg + predposs

{p 0 7}{space 1}{text:( 1)}{space 1}{space 1}{res}- prednegg + predposs = 0{p_end}

{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 1}     fetstat{col 14}{c |}      Coef.{col 26}   Std. Err.{col 38}      t{col 46}   P>|t|{col 54}     [95% Con{col 67}f. Interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 9}(1) {c |}{col 14}{res}{space 2} .0604639{col 26}{space 2} .0063125{col 37}{space 1}    9.58{col 46}{space 3}0.000{col 54}{space 4} .0477156{col 67}{space 3} .0732123
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}

{com}. //This regression gives equal weight to each study
. // TABLE 5 - Panel B/Column 2
. regress fetstat feprecision prednegg predposs [pweight = weight], vce(cluster idstudy)
{txt}(sum of wgt is 38.11478993733051)

Linear regression                               Number of obs     = {res}       641
                                                {txt}F(3, 41)          =  {res}    32.00
                                                {txt}Prob > F          = {res}    0.0000
                                                {txt}R-squared         = {res}    0.1296
                                                {txt}Root MSE          =    {res} 2.1004

{txt}{ralign 78:(Std. Err. adjusted for {res:42} clusters in idstudy)}
{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26}    Robust
{col 1}     fetstat{col 14}{c |}      Coef.{col 26}   Std. Err.{col 38}      t{col 46}   P>|t|{col 54}     [95% Con{col 67}f. Interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 1}feprecision {c |}{col 14}{res}{space 2}-.0006188{col 26}{space 2} .0009016{col 37}{space 1}   -0.69{col 46}{space 3}0.496{col 54}{space 4}-.0024396{col 67}{space 3}  .001202
{txt}{space 4}prednegg {c |}{col 14}{res}{space 2}-.0289242{col 26}{space 2} .0097438{col 37}{space 1}   -2.97{col 46}{space 3}0.005{col 54}{space 4}-.0486022{col 67}{space 3}-.0092461
{txt}{space 4}predposs {c |}{col 14}{res}{space 2} .0380316{col 26}{space 2} .0043109{col 37}{space 1}    8.82{col 46}{space 3}0.000{col 54}{space 4} .0293257{col 67}{space 3} .0467376
{txt}{space 7}_cons {c |}{col 14}{res}{space 2}-1.500027{col 26}{space 2} .2660704{col 37}{space 1}   -5.64{col 46}{space 3}0.000{col 54}{space 4}-2.037367{col 67}{space 3}-.9626872
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. lincom -prednegg + predposs

{p 0 7}{space 1}{text:( 1)}{space 1}{space 1}{res}- prednegg + predposs = 0{p_end}

{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 1}     fetstat{col 14}{c |}      Coef.{col 26}   Std. Err.{col 38}      t{col 46}   P>|t|{col 54}     [95% Con{col 67}f. Interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 9}(1) {c |}{col 14}{res}{space 2} .0669558{col 26}{space 2} .0101784{col 37}{space 1}    6.58{col 46}{space 3}0.000{col 54}{space 4} .0464001{col 67}{space 3} .0875116
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}

{com}. 
. // RANDOM EFFECTS
. 
. // Generating transformed variables for RE
. 
. metareg coefficient peerreviewed pubyear cs length midyear gdp marginal differenced etr  ///
> investment tradeopenness human popgrowth employgrowth unemployment inflation income laggeddv ///
> countryfe sehac sehet lrcase2 lrcase3 gls endog eu15 g7 eumem ///
> labourtax capitaltax othertaxes mixedtaxes overalltax predneg predpos se, wsse(se)

{txt}Meta-regression{col 55}Number of obs{col 70}= {res}    641
REML{txt} estimate of between-study variance{col 55}tau2{col 70}={res} .004458
{txt}% residual variation due to heterogeneity{col 55}I-squared_res{col 70}= {res} 68.56%
{txt}Proportion of between-study variance explained{col 55}Adj R-squared {col 70}= {res} 62.13%
{txt}Joint test for all covariates{col 55}Model F({res}36{txt},{res}604{txt}){col 70}= {res}  15.00
With{txt} Knapp-Hartung modification{col 55}Prob > F{col 70}= {res} 0.0000
{txt}{hline 17}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 1}     coefficient{col 18}{c |}      Coef.{col 30}   Std. Err.{col 42}      t{col 50}   P>|t|{col 58}     [95% Con{col 71}f. Interval]
{hline 17}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 4}peerreviewed {c |}{col 18}{res}{space 2} .0208276{col 30}{space 2} .0129336{col 41}{space 1}    1.61{col 50}{space 3}0.108{col 58}{space 4}-.0045726{col 71}{space 3} .0462278
{txt}{space 9}pubyear {c |}{col 18}{res}{space 2} .0064456{col 30}{space 2} .0026179{col 41}{space 1}    2.46{col 50}{space 3}0.014{col 58}{space 4} .0013043{col 71}{space 3}  .011587
{txt}{space 14}cs {c |}{col 18}{res}{space 2} .0732605{col 30}{space 2} .0348075{col 41}{space 1}    2.10{col 50}{space 3}0.036{col 58}{space 4}  .004902{col 71}{space 3}  .141619
{txt}{space 10}length {c |}{col 18}{res}{space 2} .0002601{col 30}{space 2} .0014532{col 41}{space 1}    0.18{col 50}{space 3}0.858{col 58}{space 4}-.0025938{col 71}{space 3} .0031141
{txt}{space 9}midyear {c |}{col 18}{res}{space 2}-.0047186{col 30}{space 2} .0028915{col 41}{space 1}   -1.63{col 50}{space 3}0.103{col 58}{space 4}-.0103972{col 71}{space 3}   .00096
{txt}{space 13}gdp {c |}{col 18}{res}{space 2} .0661244{col 30}{space 2} .0184299{col 41}{space 1}    3.59{col 50}{space 3}0.000{col 58}{space 4} .0299299{col 71}{space 3} .1023189
{txt}{space 8}marginal {c |}{col 18}{res}{space 2} .0453306{col 30}{space 2} .0250162{col 41}{space 1}    1.81{col 50}{space 3}0.070{col 58}{space 4}-.0037988{col 71}{space 3}   .09446
{txt}{space 5}differenced {c |}{col 18}{res}{space 2}-.0443335{col 30}{space 2} .0256176{col 41}{space 1}   -1.73{col 50}{space 3}0.084{col 58}{space 4} -.094644{col 71}{space 3} .0059769
{txt}{space 13}etr {c |}{col 18}{res}{space 2} .0364516{col 30}{space 2} .0250249{col 41}{space 1}    1.46{col 50}{space 3}0.146{col 58}{space 4}-.0126949{col 71}{space 3}  .085598
{txt}{space 6}investment {c |}{col 18}{res}{space 2} .0066733{col 30}{space 2} .0143768{col 41}{space 1}    0.46{col 50}{space 3}0.643{col 58}{space 4}-.0215613{col 71}{space 3}  .034908
{txt}{space 3}tradeopenness {c |}{col 18}{res}{space 2}-.0040907{col 30}{space 2} .0163533{col 41}{space 1}   -0.25{col 50}{space 3}0.803{col 58}{space 4} -.036207{col 71}{space 3} .0280255
{txt}{space 11}human {c |}{col 18}{res}{space 2}-.0210297{col 30}{space 2} .0138299{col 41}{space 1}   -1.52{col 50}{space 3}0.129{col 58}{space 4}-.0481902{col 71}{space 3} .0061308
{txt}{space 7}popgrowth {c |}{col 18}{res}{space 2}-.0419907{col 30}{space 2} .0175187{col 41}{space 1}   -2.40{col 50}{space 3}0.017{col 58}{space 4}-.0763956{col 71}{space 3}-.0075858
{txt}{space 4}employgrowth {c |}{col 18}{res}{space 2} .0220656{col 30}{space 2} .0178401{col 41}{space 1}    1.24{col 50}{space 3}0.217{col 58}{space 4}-.0129706{col 71}{space 3} .0571018
{txt}unemploymentrate {c |}{col 18}{res}{space 2} .0774464{col 30}{space 2} .0194147{col 41}{space 1}    3.99{col 50}{space 3}0.000{col 58}{space 4} .0393179{col 71}{space 3} .1155749
{txt}{space 7}inflation {c |}{col 18}{res}{space 2}-.0352781{col 30}{space 2} .0171199{col 41}{space 1}   -2.06{col 50}{space 3}0.040{col 58}{space 4}-.0688998{col 71}{space 3}-.0016564
{txt}{space 10}income {c |}{col 18}{res}{space 2} .0571603{col 30}{space 2} .0179417{col 41}{space 1}    3.19{col 50}{space 3}0.002{col 58}{space 4} .0219246{col 71}{space 3} .0923959
{txt}{space 8}laggeddv {c |}{col 18}{res}{space 2} .0064412{col 30}{space 2} .0216167{col 41}{space 1}    0.30{col 50}{space 3}0.766{col 58}{space 4}-.0360118{col 71}{space 3} .0488941
{txt}{space 7}countryfe {c |}{col 18}{res}{space 2}-.0317365{col 30}{space 2} .0159844{col 41}{space 1}   -1.99{col 50}{space 3}0.048{col 58}{space 4}-.0631282{col 71}{space 3}-.0003448
{txt}{space 11}sehac {c |}{col 18}{res}{space 2} .0586019{col 30}{space 2} .0240402{col 41}{space 1}    2.44{col 50}{space 3}0.015{col 58}{space 4} .0113893{col 71}{space 3} .1058144
{txt}{space 11}sehet {c |}{col 18}{res}{space 2} .0245161{col 30}{space 2} .0129482{col 41}{space 1}    1.89{col 50}{space 3}0.059{col 58}{space 4}-.0009129{col 71}{space 3} .0499452
{txt}{space 9}lrcase2 {c |}{col 18}{res}{space 2} .0904042{col 30}{space 2}  .030275{col 41}{space 1}    2.99{col 50}{space 3}0.003{col 58}{space 4} .0309471{col 71}{space 3} .1498614
{txt}{space 9}lrcase3 {c |}{col 18}{res}{space 2}-.0675386{col 30}{space 2} .0147489{col 41}{space 1}   -4.58{col 50}{space 3}0.000{col 58}{space 4}-.0965038{col 71}{space 3}-.0385733
{txt}{space 13}gls {c |}{col 18}{res}{space 2}-.0444032{col 30}{space 2} .0223131{col 41}{space 1}   -1.99{col 50}{space 3}0.047{col 58}{space 4}-.0882238{col 71}{space 3}-.0005825
{txt}{space 11}endog {c |}{col 18}{res}{space 2}-.0043295{col 30}{space 2} .0126907{col 41}{space 1}   -0.34{col 50}{space 3}0.733{col 58}{space 4}-.0292527{col 71}{space 3} .0205937
{txt}{space 12}eu15 {c |}{col 18}{res}{space 2} .1268459{col 30}{space 2} .0285493{col 41}{space 1}    4.44{col 50}{space 3}0.000{col 58}{space 4}  .070778{col 71}{space 3} .1829137
{txt}{space 14}g7 {c |}{col 18}{res}{space 2} .1358728{col 30}{space 2} .0333582{col 41}{space 1}    4.07{col 50}{space 3}0.000{col 58}{space 4} .0703606{col 71}{space 3}  .201385
{txt}{space 11}eumem {c |}{col 18}{res}{space 2} .0154951{col 30}{space 2} .0616331{col 41}{space 1}    0.25{col 50}{space 3}0.802{col 58}{space 4}-.1055461{col 71}{space 3} .1365364
{txt}{space 7}labourtax {c |}{col 18}{res}{space 2}-.0382743{col 30}{space 2} .0190231{col 41}{space 1}   -2.01{col 50}{space 3}0.045{col 58}{space 4}-.0756338{col 71}{space 3}-.0009148
{txt}{space 6}capitaltax {c |}{col 18}{res}{space 2} -.002245{col 30}{space 2} .0192118{col 41}{space 1}   -0.12{col 50}{space 3}0.907{col 58}{space 4} -.039975{col 71}{space 3}  .035485
{txt}{space 6}othertaxes {c |}{col 18}{res}{space 2} .1844773{col 30}{space 2} .2914955{col 41}{space 1}    0.63{col 50}{space 3}0.527{col 58}{space 4}-.3879905{col 71}{space 3} .7569451
{txt}{space 6}mixedtaxes {c |}{col 18}{res}{space 2} -.087798{col 30}{space 2} .0200726{col 41}{space 1}   -4.37{col 50}{space 3}0.000{col 58}{space 4}-.1272186{col 71}{space 3}-.0483775
{txt}{space 6}overalltax {c |}{col 18}{res}{space 2} -.045274{col 30}{space 2} .0216387{col 41}{space 1}   -2.09{col 50}{space 3}0.037{col 58}{space 4}-.0877702{col 71}{space 3}-.0027778
{txt}{space 9}predneg {c |}{col 18}{res}{space 2}-.0708786{col 30}{space 2} .0146457{col 41}{space 1}   -4.84{col 50}{space 3}0.000{col 58}{space 4}-.0996412{col 71}{space 3} -.042116
{txt}{space 9}predpos {c |}{col 18}{res}{space 2} .0289162{col 30}{space 2} .0236641{col 41}{space 1}    1.22{col 50}{space 3}0.222{col 58}{space 4}-.0175577{col 71}{space 3}   .07539
{txt}{space 14}se {c |}{col 18}{res}{space 2}-.7561872{col 30}{space 2} .1194345{col 41}{space 1}   -6.33{col 50}{space 3}0.000{col 58}{space 4}-.9907446{col 71}{space 3}-.5216297
{txt}{space 11}_cons {c |}{col 18}{res}{space 2}-.0088096{col 30}{space 2} .1079974{col 41}{space 1}   -0.08{col 50}{space 3}0.935{col 58}{space 4}-.2209057{col 71}{space 3} .2032864
{txt}{hline 17}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}

{com}. scalar tau2 = e(tau2)
{txt}
{com}. gen revar = se^2 + tau2
{txt}
{com}. gen rese = sqrt(revar)
{txt}
{com}. gen reprecision = 1/rese
{txt}
{com}. gen retstat = coefficient/rese
{txt}
{com}. gen repubbias = se/rese
{txt}
{com}. replace prednegg = predneg/rese
{txt}(146 real changes made)

{com}. replace predposs = predpos/rese
{txt}(38 real changes made)

{com}. 
. // TABLE 5/PANEL A
. // NO CORRECTION FOR PUBLICATION BIAS
. //This regression gives equal weight to each estimate
. // TABLE 5 - Panel A/Column 3
. regress retstat reprecision prednegg predposs, noc vce(cluster idstudy)

{txt}Linear regression                               Number of obs     = {res}       641
                                                {txt}F(3, 41)          =  {res}     7.36
                                                {txt}Prob > F          = {res}    0.0005
                                                {txt}R-squared         = {res}    0.3356
                                                {txt}Root MSE          =    {res} 1.3147

{txt}{ralign 78:(Std. Err. adjusted for {res:42} clusters in idstudy)}
{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26}    Robust
{col 1}     retstat{col 14}{c |}      Coef.{col 26}   Std. Err.{col 38}      t{col 46}   P>|t|{col 54}     [95% Con{col 67}f. Interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 1}reprecision {c |}{col 14}{res}{space 2}-.0814517{col 26}{space 2}  .019084{col 37}{space 1}   -4.27{col 46}{space 3}0.000{col 54}{space 4}-.1199926{col 67}{space 3}-.0429107
{txt}{space 4}prednegg {c |}{col 14}{res}{space 2}-.0238345{col 26}{space 2} .0517301{col 37}{space 1}   -0.46{col 46}{space 3}0.647{col 54}{space 4}-.1283056{col 67}{space 3} .0806366
{txt}{space 4}predposs {c |}{col 14}{res}{space 2} .0840646{col 26}{space 2} .0260093{col 37}{space 1}    3.23{col 46}{space 3}0.002{col 54}{space 4} .0315377{col 67}{space 3} .1365915
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. lincom -prednegg + predposs

{p 0 7}{space 1}{text:( 1)}{space 1}{space 1}{res}- prednegg + predposs = 0{p_end}

{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 1}     retstat{col 14}{c |}      Coef.{col 26}   Std. Err.{col 38}      t{col 46}   P>|t|{col 54}     [95% Con{col 67}f. Interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 9}(1) {c |}{col 14}{res}{space 2}  .107899{col 26}{space 2} .0513519{col 37}{space 1}    2.10{col 46}{space 3}0.042{col 54}{space 4} .0041918{col 67}{space 3} .2116062
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}

{com}. //This regression gives equal weight to each study
. // TABLE 5 - Panel A/Column 4
. regress retstat reprecision prednegg predposs [pweight = weight], noc vce(cluster idstudy)
{txt}(sum of wgt is 38.11478993733051)

Linear regression                               Number of obs     = {res}       641
                                                {txt}F(3, 41)          =  {res}     8.48
                                                {txt}Prob > F          = {res}    0.0002
                                                {txt}R-squared         = {res}    0.3197
                                                {txt}Root MSE          =    {res} 1.2728

{txt}{ralign 78:(Std. Err. adjusted for {res:42} clusters in idstudy)}
{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26}    Robust
{col 1}     retstat{col 14}{c |}      Coef.{col 26}   Std. Err.{col 38}      t{col 46}   P>|t|{col 54}     [95% Con{col 67}f. Interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 1}reprecision {c |}{col 14}{res}{space 2}-.0638267{col 26}{space 2} .0156501{col 37}{space 1}   -4.08{col 46}{space 3}0.000{col 54}{space 4}-.0954327{col 67}{space 3}-.0322206
{txt}{space 4}prednegg {c |}{col 14}{res}{space 2}  -.06769{col 26}{space 2} .0463521{col 37}{space 1}   -1.46{col 46}{space 3}0.152{col 54}{space 4}-.1612999{col 67}{space 3}   .02592
{txt}{space 4}predposs {c |}{col 14}{res}{space 2} .0691709{col 26}{space 2}  .024497{col 37}{space 1}    2.82{col 46}{space 3}0.007{col 54}{space 4} .0196981{col 67}{space 3} .1186437
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. lincom -prednegg + predposs

{p 0 7}{space 1}{text:( 1)}{space 1}{space 1}{res}- prednegg + predposs = 0{p_end}

{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 1}     retstat{col 14}{c |}      Coef.{col 26}   Std. Err.{col 38}      t{col 46}   P>|t|{col 54}     [95% Con{col 67}f. Interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 9}(1) {c |}{col 14}{res}{space 2} .1368609{col 26}{space 2} .0471106{col 37}{space 1}    2.91{col 46}{space 3}0.006{col 54}{space 4} .0417191{col 67}{space 3} .2320027
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}

{com}. 
. // TABLE 5/PANEL B
. // CORRECTION FOR PUBLICATION BIAS
. //This regression gives equal weight to each estimate
. // TABLE 5 - Panel B/Column 3
. regress retstat reprecision repubbias prednegg predposs, noc vce(cluster idstudy)

{txt}Linear regression                               Number of obs     = {res}       641
                                                {txt}F(4, 41)          =  {res}    10.06
                                                {txt}Prob > F          = {res}    0.0000
                                                {txt}R-squared         = {res}    0.4045
                                                {txt}Root MSE          =    {res} 1.2456

{txt}{ralign 78:(Std. Err. adjusted for {res:42} clusters in idstudy)}
{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26}    Robust
{col 1}     retstat{col 14}{c |}      Coef.{col 26}   Std. Err.{col 38}      t{col 46}   P>|t|{col 54}     [95% Con{col 67}f. Interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 1}reprecision {c |}{col 14}{res}{space 2}-.0366687{col 26}{space 2} .0205912{col 37}{space 1}   -1.78{col 46}{space 3}0.082{col 54}{space 4}-.0782536{col 67}{space 3} .0049161
{txt}{space 3}repubbias {c |}{col 14}{res}{space 2}-.9035349{col 26}{space 2} .2810398{col 37}{space 1}   -3.21{col 46}{space 3}0.003{col 54}{space 4}-1.471106{col 67}{space 3}-.3359635
{txt}{space 4}prednegg {c |}{col 14}{res}{space 2}-.0318917{col 26}{space 2} .0406236{col 37}{space 1}   -0.79{col 46}{space 3}0.437{col 54}{space 4}-.1139326{col 67}{space 3} .0501493
{txt}{space 4}predposs {c |}{col 14}{res}{space 2} .0828782{col 26}{space 2} .0291428{col 37}{space 1}    2.84{col 46}{space 3}0.007{col 54}{space 4}  .024023{col 67}{space 3} .1417333
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. lincom -prednegg + predposs

{p 0 7}{space 1}{text:( 1)}{space 1}{space 1}{res}- prednegg + predposs = 0{p_end}

{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 1}     retstat{col 14}{c |}      Coef.{col 26}   Std. Err.{col 38}      t{col 46}   P>|t|{col 54}     [95% Con{col 67}f. Interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 9}(1) {c |}{col 14}{res}{space 2} .1147698{col 26}{space 2} .0422721{col 37}{space 1}    2.72{col 46}{space 3}0.010{col 54}{space 4} .0293995{col 67}{space 3} .2001401
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}

{com}. //This regression gives equal weight to each study
. // TABLE 5 - Panel B/Column 4
. regress retstat reprecision repubbias prednegg predposs [pweight = weight], noc vce(cluster idstudy)
{txt}(sum of wgt is 38.11478993733051)

Linear regression                               Number of obs     = {res}       641
                                                {txt}F(4, 41)          =  {res}    10.87
                                                {txt}Prob > F          = {res}    0.0000
                                                {txt}R-squared         = {res}    0.4041
                                                {txt}Root MSE          =    {res} 1.1922

{txt}{ralign 78:(Std. Err. adjusted for {res:42} clusters in idstudy)}
{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26}    Robust
{col 1}     retstat{col 14}{c |}      Coef.{col 26}   Std. Err.{col 38}      t{col 46}   P>|t|{col 54}     [95% Con{col 67}f. Interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 1}reprecision {c |}{col 14}{res}{space 2}-.0248064{col 26}{space 2} .0138861{col 37}{space 1}   -1.79{col 46}{space 3}0.081{col 54}{space 4}  -.05285{col 67}{space 3} .0032371
{txt}{space 3}repubbias {c |}{col 14}{res}{space 2}-.9904061{col 26}{space 2} .2121476{col 37}{space 1}   -4.67{col 46}{space 3}0.000{col 54}{space 4}-1.418847{col 67}{space 3}-.5619653
{txt}{space 4}prednegg {c |}{col 14}{res}{space 2}-.0563761{col 26}{space 2} .0384873{col 37}{space 1}   -1.46{col 46}{space 3}0.151{col 54}{space 4}-.1341027{col 67}{space 3} .0213505
{txt}{space 4}predposs {c |}{col 14}{res}{space 2} .0635695{col 26}{space 2} .0205068{col 37}{space 1}    3.10{col 46}{space 3}0.003{col 54}{space 4} .0221553{col 67}{space 3} .1049838
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. lincom -prednegg + predposs

{p 0 7}{space 1}{text:( 1)}{space 1}{space 1}{res}- prednegg + predposs = 0{p_end}

{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 1}     retstat{col 14}{c |}      Coef.{col 26}   Std. Err.{col 38}      t{col 46}   P>|t|{col 54}     [95% Con{col 67}f. Interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 9}(1) {c |}{col 14}{res}{space 2} .1199457{col 26}{space 2} .0384554{col 37}{space 1}    3.12{col 46}{space 3}0.003{col 54}{space 4} .0422835{col 67}{space 3} .1976079
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}

{com}. 
. *----------------------------------------------------------------*
. *   APPENDIX 5: Fixed Effects                                    *
. *----------------------------------------------------------------*
. // Determining the weight of individual studies (Fixed Effects)
. preserve
{txt}
{com}. gen fevarR = se^2
{txt}
{com}. gen feweightR = 1/fevarR
{txt}
{com}. gen r = coefficient
{txt}
{com}. gen rR = r*feweightR
{txt}
{com}. collapse (sum) rR feweightR, by (idstudy)
{txt}
{com}. gen avgR = rR/feweightR
{txt}
{com}. gen avgsterrR = sqrt(1/feweightR)
{txt}
{com}. sort idstudy
{txt}
{com}. metan avgR avgsterrR, fixed lcols(id) nobox

{txt}{col 12}Study{col 22}|{col 24}{col 28}ES{col 34}[95% Conf. Interval]{col 59}% Weight
---------------------+---------------------------------------------------
1{col 22}|{res}  0.061{col 35} -0.019     0.142{col 60}  0.01
{txt}2{col 22}|{res} -0.139{col 35} -0.200    -0.077{col 60}  0.02
{txt}3{col 22}|{res} -0.037{col 35} -0.133     0.059{col 60}  0.01
{txt}4{col 22}|{res} -0.003{col 35} -0.029     0.023{col 60}  0.10
{txt}5{col 22}|{res} -0.173{col 35} -0.442     0.097{col 60}  0.00
{txt}6{col 22}|{res} -0.071{col 35} -0.193     0.050{col 60}  0.00
{txt}7{col 22}|{res} -0.136{col 35} -0.216    -0.057{col 60}  0.01
{txt}8{col 22}|{res} -0.008{col 35} -0.018     0.002{col 60}  0.67
{txt}9{col 22}|{res} -0.052{col 35} -0.080    -0.023{col 60}  0.08
{txt}10{col 22}|{res} -0.218{col 35} -0.257    -0.178{col 60}  0.04
{txt}11{col 22}|{res} -0.260{col 35} -0.281    -0.240{col 60}  0.16
{txt}12{col 22}|{res} -0.076{col 35} -0.101    -0.051{col 60}  0.11
{txt}13{col 22}|{res}  0.006{col 35} -0.034     0.047{col 60}  0.04
{txt}14{col 22}|{res} -0.092{col 35} -0.115    -0.070{col 60}  0.13
{txt}15{col 22}|{res} -0.432{col 35} -0.472    -0.393{col 60}  0.04
{txt}16{col 22}|{res}  0.020{col 35} -0.001     0.041{col 60}  0.15
{txt}17{col 22}|{res}  0.004{col 35} -0.012     0.021{col 60}  0.26
{txt}18{col 22}|{res} -0.026{col 35} -0.037    -0.015{col 60}  0.56
{txt}19{col 22}|{res} -0.030{col 35} -0.057    -0.002{col 60}  0.09
{txt}20{col 22}|{res} -0.030{col 35} -0.049    -0.011{col 60}  0.19
{txt}21{col 22}|{res} -0.149{col 35} -0.174    -0.125{col 60}  0.11
{txt}22{col 22}|{res}  0.006{col 35}  0.002     0.010{col 60}  4.04
{txt}23{col 22}|{res} -0.118{col 35} -0.136    -0.100{col 60}  0.21
{txt}24{col 22}|{res} -0.034{col 35} -0.037    -0.030{col 60}  6.96
{txt}25{col 22}|{res} -0.118{col 35} -0.136    -0.100{col 60}  0.21
{txt}26{col 22}|{res} -0.002{col 35} -0.003    -0.001{col 60} 81.39
{txt}27{col 22}|{res} -0.042{col 35} -0.068    -0.015{col 60}  0.09
{txt}28{col 22}|{res} -0.178{col 35} -0.205    -0.151{col 60}  0.09
{txt}29{col 22}|{res} -0.028{col 35} -0.055    -0.001{col 60}  0.09
{txt}30{col 22}|{res} -0.295{col 35} -0.345    -0.245{col 60}  0.03
{txt}31{col 22}|{res} -0.004{col 35} -0.015     0.008{col 60}  0.49
{txt}32{col 22}|{res} -0.325{col 35} -0.596    -0.053{col 60}  0.00
{txt}33{col 22}|{res} -0.123{col 35} -0.256     0.010{col 60}  0.00
{txt}34{col 22}|{res} -0.012{col 35} -0.019    -0.005{col 60}  1.42
{txt}35{col 22}|{res} -0.006{col 35} -0.014     0.001{col 60}  1.21
{txt}36{col 22}|{res} -0.029{col 35} -0.042    -0.016{col 60}  0.39
{txt}37{col 22}|{res}  0.005{col 35} -0.013     0.022{col 60}  0.23
{txt}38{col 22}|{res} -0.238{col 35} -0.254    -0.222{col 60}  0.25
{txt}39{col 22}|{res} -0.399{col 35} -3.798     3.000{col 60}  0.00
{txt}40{col 22}|{res} -0.067{col 35} -0.103    -0.031{col 60}  0.05
{txt}41{col 22}|{res} -0.285{col 35} -0.338    -0.232{col 60}  0.02
{txt}42{col 22}|{res} -0.274{col 35} -0.328    -0.220{col 60}  0.02
{txt}---------------------+---------------------------------------------------
I-V pooled ES{col 22}|{res} -0.007{col 35} -0.008    -0.006{col 60}100.00
{txt}---------------------+---------------------------------------------------

  Heterogeneity chi-squared = {res}3440.44{txt} (d.f. = {res}41{txt}) p = {res}0.000
{txt}  I-squared (variation in ES attributable to heterogeneity) ={res}  98.8%

{txt}  Test of ES=0 : z= {res} 16.54{txt} p = {res}0.000
{txt}
{com}. gen studyweight = _WT
{txt}
{com}. summ studyweight, detail

                         {txt}studyweight
{hline 61}
      Percentiles      Smallest
 1%    {res} 5.86e-06       5.86e-06
{txt} 5%    {res} .0009325       .0009184
{txt}10%    {res} .0045824       .0009325       {txt}Obs         {res}         42
{txt}25%    {res}  .024316       .0038305       {txt}Sum of Wgt. {res}         42

{txt}50%    {res} .0976544                      {txt}Mean          {res} 2.380952
                        {txt}Largest       Std. Dev.     {res} 12.54825
{txt}75%    {res} .2529721       1.418938
{txt}90%    {res} 1.206384       4.042448       {txt}Variance      {res} 157.4587
{txt}95%    {res} 4.042448       6.956644       {txt}Skewness      {res} 6.157658
{txt}99%    {res} 81.38846       81.38846       {txt}Kurtosis      {res} 39.26789
{txt}
{com}. gsort -studyweight
{txt}
{com}. // NOTE: The top 3 studies account for approximately 92% of the weight! 
. restore
{txt}
{com}. 
. *-----------------------------------------------------------------*
. *   APPENDIX 5: Random Effects                                    *
. *-----------------------------------------------------------------*
. // Determining the weight of individual studies (Random Effects)
. gen reweightR = 1/revar
{txt}
{com}. gen r = coefficient
{txt}
{com}. gen rR = r*reweightR
{txt}
{com}. collapse (sum) rR reweightR, by (idstudy)
{txt}
{com}. gen avgR = rR/reweightR
{txt}
{com}. gen avgsterrR = sqrt(1/reweightR)
{txt}
{com}. sort idstudy
{txt}
{com}. metan avgR avgsterrR, random lcols(id) nobox

{txt}{col 12}Study{col 22}|{col 24}{col 28}ES{col 34}[95% Conf. Interval]{col 59}% Weight
---------------------+---------------------------------------------------
1{col 22}|{res}  0.044{col 35} -0.071     0.158{col 60}  2.05
{txt}2{col 22}|{res} -0.175{col 35} -0.260    -0.089{col 60}  2.34
{txt}3{col 22}|{res} -0.045{col 35} -0.165     0.075{col 60}  2.00
{txt}4{col 22}|{res} -0.001{col 35} -0.081     0.079{col 60}  2.39
{txt}5{col 22}|{res} -0.165{col 35} -0.448     0.117{col 60}  0.86
{txt}6{col 22}|{res} -0.051{col 35} -0.199     0.096{col 60}  1.74
{txt}7{col 22}|{res} -0.143{col 35} -0.242    -0.044{col 60}  2.21
{txt}8{col 22}|{res} -0.018{col 35} -0.047     0.010{col 60}  2.77
{txt}9{col 22}|{res} -0.053{col 35} -0.087    -0.019{col 60}  2.74
{txt}10{col 22}|{res} -0.223{col 35} -0.294    -0.152{col 60}  2.47
{txt}11{col 22}|{res} -0.284{col 35} -0.322    -0.246{col 60}  2.72
{txt}12{col 22}|{res} -0.095{col 35} -0.142    -0.048{col 60}  2.66
{txt}13{col 22}|{res}  0.010{col 35} -0.076     0.096{col 60}  2.33
{txt}14{col 22}|{res} -0.090{col 35} -0.139    -0.041{col 60}  2.65
{txt}15{col 22}|{res} -0.429{col 35} -0.482    -0.376{col 60}  2.62
{txt}16{col 22}|{res}  0.017{col 35} -0.032     0.065{col 60}  2.65
{txt}17{col 22}|{res}  0.017{col 35} -0.042     0.076{col 60}  2.57
{txt}18{col 22}|{res} -0.035{col 35} -0.072     0.002{col 60}  2.72
{txt}19{col 22}|{res} -0.042{col 35} -0.100     0.017{col 60}  2.58
{txt}20{col 22}|{res} -0.076{col 35} -0.134    -0.018{col 60}  2.58
{txt}21{col 22}|{res} -0.146{col 35} -0.180    -0.113{col 60}  2.74
{txt}22{col 22}|{res}  0.007{col 35} -0.036     0.049{col 60}  2.69
{txt}23{col 22}|{res} -0.123{col 35} -0.162    -0.084{col 60}  2.71
{txt}24{col 22}|{res} -0.029{col 35} -0.048    -0.010{col 60}  2.80
{txt}25{col 22}|{res} -0.124{col 35} -0.163    -0.085{col 60}  2.71
{txt}26{col 22}|{res} -0.025{col 35} -0.053     0.004{col 60}  2.77
{txt}27{col 22}|{res} -0.042{col 35} -0.138     0.054{col 60}  2.24
{txt}28{col 22}|{res} -0.188{col 35} -0.226    -0.150{col 60}  2.72
{txt}29{col 22}|{res} -0.052{col 35} -0.099    -0.004{col 60}  2.66
{txt}30{col 22}|{res} -0.288{col 35} -0.347    -0.230{col 60}  2.58
{txt}31{col 22}|{res} -0.006{col 35} -0.047     0.035{col 60}  2.70
{txt}32{col 22}|{res} -0.324{col 35} -0.607    -0.041{col 60}  0.86
{txt}33{col 22}|{res} -0.133{col 35} -0.279     0.013{col 60}  1.75
{txt}34{col 22}|{res} -0.012{col 35} -0.104     0.081{col 60}  2.27
{txt}35{col 22}|{res}  0.002{col 35} -0.034     0.038{col 60}  2.73
{txt}36{col 22}|{res} -0.039{col 35} -0.075    -0.004{col 60}  2.73
{txt}37{col 22}|{res}  0.000{col 35} -0.056     0.057{col 60}  2.59
{txt}38{col 22}|{res} -0.202{col 35} -0.233    -0.170{col 60}  2.75
{txt}39{col 22}|{res} -0.399{col 35} -3.801     3.003{col 60}  0.01
{txt}40{col 22}|{res} -0.060{col 35} -0.131     0.010{col 60}  2.48
{txt}41{col 22}|{res} -0.273{col 35} -0.343    -0.204{col 60}  2.49
{txt}42{col 22}|{res} -0.255{col 35} -0.339    -0.171{col 60}  2.36
{txt}---------------------+---------------------------------------------------
D+L pooled ES{col 22}|{res} -0.097{col 35} -0.129    -0.066{col 60}100.00
{txt}---------------------+---------------------------------------------------

  Heterogeneity chi-squared = {res}684.31{txt} (d.f. = {res}41{txt}) p = {res}0.000
{txt}  I-squared (variation in ES attributable to heterogeneity) ={res}  94.0%
{txt}  Estimate of between-study variance Tau-squared = {res} 0.0091

{txt}  Test of ES=0 : z= {res}  6.06{txt} p = {res}0.000
{txt}
{com}. gen studyweight = _WT
{txt}
{com}. summ studyweight, detail

                         {txt}studyweight
{hline 61}
      Percentiles      Smallest
 1%    {res} .0084948       .0084948
{txt} 5%    {res} .8601947       .8563909
{txt}10%    {res} 1.752454       .8601947       {txt}Obs         {res}         42
{txt}25%    {res} 2.333486       1.743431       {txt}Sum of Wgt. {res}         42

{txt}50%    {res} 2.587937                      {txt}Mean          {res} 2.380952
                        {txt}Largest       Std. Dev.     {res} .5812799
{txt}75%    {res} 2.717981        2.75427
{txt}90%    {res} 2.743071       2.765758       {txt}Variance      {res} .3378863
{txt}95%    {res} 2.765758       2.767675       {txt}Skewness      {res}-2.539957
{txt}99%    {res} 2.801705       2.801705       {txt}Kurtosis      {res} 9.424142
{txt}
{com}. gsort -studyweight
{txt}
{com}. // NOTE: Random effects very evenly (too evenly?) weighted
. 
. log close
      {txt}name:  {res}<unnamed>
       {txt}log:  {res}\\file\UsersW$\wrr15\Home\My Documents\My Files\TAXES AND ECONOMIC GROWTH IN OECD COUNTRIES - A META-ANALYSIS\SUBMISSION TO PFR\DATA AND PROGRAMS\Part1 Results(20200102).smcl
  {txt}log type:  {res}smcl
 {txt}closed on:  {res} 3 Jan 2020, 07:50:08
{txt}{.-}
{smcl}
{txt}{sf}{ul off}